Provide better error message when users try to run incremental scans on a history table

[kbendick]

It's not possible to run an incremental scan on a history metadata table.

Occasionally, users will swap back and forth between doing an incremental scan and querying the history table and will forget to unset the incremental versions, and the error message is not very helpful.

Here's an example:

spark.read.format("iceberg")
  .option("start-snapshot-id", "4718831532792232068")
  .option("end-snapshot-id", "4044746363673034999")
  .load("namespace.table.history")
  .show()

The error message

Message: Incremental scan is not supported

It would likely be much more helpful if it said something like Incremental scan is not supported in history metadata table or more generally Incremental scan is not supported in <actual class type>.

[kbendick]

This probably only comes up in the MetadataTable class, so not sure if it would be easier to update the base class or just that one place. I will take a look and submit a patch.